Background: The American Cancer Society-National Prostate Cancer Detection Project (ACS-NPCDP) is a multidisciplinary evaluation of early prostate cancer detection interventions. This report summarizes the experience of the investigators to date and describes the overall and relative performance of the different detection modalities studied in this project.
Methods: Two thousand nine hundred ninety-nine men aged 55 to 70 years at entry who were not already under evaluation for prostate cancer were recruited to participate in up to 5 annual examinations by prostate specific antigen (PSA), digital rectal examination (DRE), and transrectal ultrasound (TRUS). In the course of 5 years of intervention, ACS-NPCDP investigators have completed 9937 examinations, recommended 1215 biopsies, and detected 203 cancers.
Results: Loss to cohort follow-up was greatest in the first year. Overall, TRUS led to twice the number of recommendations for biopsy compared with DRE (8.9% versus 4.4%). Elevated PSA was observed in 13.0% of 9535 measurements performed. The overall cancer detection rate declined significantly during the five years of intervention. Detection was significantly associated with age and symptom status at entry. DRE had lower sensitivity compared with TRUS or PSA, particularly in later years of follow-up. The specificity of TRUS was lower than that for DRE. PSA was elevated in 69.2% of examinations that led to cancer detection, compared with only 10.9% when cancer was not found. PSA level, PSA density, and PSA change were all related to the presence of cancer. Less than 6% of the cancers detected in this study were clinically advanced at the time of diagnosis.
Conclusions: These data quantify the yield of early cancer detection that may be expected when PSA, DRE, and TRUS are used in populations comparable to the men participating in the ACS-NPCDP. Continued follow-up and further research is needed to assess whether men receiving early prostate cancer interventions benefit as a result.